smart speakers and connected cars
This month, Kevin Ryan considers how voice-controlled devices and more listener control will affect radio
This month, Kevin Ryan considers how voice-controlled devices and more listener control will affect radio, looks at DB and DAB+ news and asks why international broadcasters struggle to deliver good DRM signals.
The other Sunday I was only ‘half-listening’ to the football on BBC Radio 5 Live, when the station identification caught my attention, announcing that BBC Radio 5 Live was available on its usual outlets and that these now also included smart speakers.
This refers to the growing use of what is now being collectively called smart speakers or voice assistants. However, we have come to know them by names like the Amazon Alexa and Google Home.
Radio stations are worried about becoming irrelevant and losing young listeners to whoever has the content they want. Last month, both the BBC and Global stated that they wished to hang on to FM.
This notwithstanding, you have to wonder whether even DAB is still the ‘brave new future’ it once was or whether it too will get squeezed out by a newer, more popular, technology.
I don’t have a smart speaker, but Alexa is on my Amazon Fire tablet computer. The press releases I found on this topic covered the BBC launching its first full-voice app for voice-controlled smart speakers.
Moreover, many of them concerned the Amazon Alexa Skill for Echo devices and previewed other smart speakers coming in the future.
A ‘skill’ is another name for a ‘plugin’ or an ‘extension’.
The BBC skill is supposed to provide access to the BBC’s full range of live radio stations, including all local, national and international radio as well as the BBC’s full range of podcasts.
The BBC will also launch brand new voice experiences exclusively for smart speakers through the service in the near future. You can download the Alexa skills from the Amazon store.
To use the skill, audiences simply need to ask for the BBC radio station (Fig. 1) or podcast they want. They are then taken straight to the live show or the latest episode of a podcast. Here, listeners are able to ask for 'previous' episodes in the full back-catalogue. Last but not least, radio consumers can 'play', 'pause' and 'resume' everything they are listening to.
I looked in the Amazon skills store and found the big names like the BBC, Capital, Heart, Absolute, RTE as well as the (more useful) Radioplayer and the myTuner radio player (Fig. 2).
Will it catch on? I don’t mind telling you that I feel a bit self-conscious talking to my Amazon Fire. Nevertheless, I am sure this is aimed at the younger generation who might adopt this technology more readily.
The skill did work on my Fire tablet at the second attempt. I asked Alexa to play BBC Radio Berkshire but instead, I got BBC Radio Essex. When I was speaking more slowly, Alexa found the station and I think it might be due to how the Americans expect you to pronounce Berkshire!
I guess that there must be some ‘training’ of the software going on in the background. The more interesting thing was that the stations were found on the BBC from the TuneIn skill, rather than the main BBC one.
The car radio gave the broadcasters a captive audience until things like CD and MP3 players arrived.
Now the smart speaker technology is appearing in automobiles too.
The use of voice commands in cars is not, in fact, a new technology, I can already talk to my car with a limited range of commands, mainly in connection with my mobile phone.
Amazon seems to be leading this field at the moment with simple skills like sounding the horn to tell you where your vehicle is parked.
I mention this because it is possible that, very soon, radio will not be the ‘default audio option, on your car’s dashboard. The big broadcasters will create voice-controlled apps to stay prominent. The smaller stations, however, might be much harder to find.
I am not sure what the rise of voice controlled devices will mean for radio in the future. However, one thing is certain: These devices are encouraging the BBC, Global and Bauer to invest heavily in the technology.
A few news items caught my eye during the month. Sound Digital, the second national commercial multiplex, is advertising for broadcasters to add to the 19 already on the multiplex. The original website has disappeared and Arqiva now hosts what SDL information there is.
SDL is also expanding its coverage in the West Country, East England, South Wales and North East Scotland. The new transmitter sites are not known just yet.
London Greek Radio (LGR, Fig. 3) expanded to Glasgow on the trial-minimux in April in DAB+.
A ‘minimux’ is the term coined for a multiplex used by a small-scale, low-power, DAB provider such as LGR.
This station is also on minimuxes in London, Birmingham and Manchester. It is interesting that their website affords more prominence to their FM frequency in London. There is more information at this URL:
Manchester Key 103 is becoming a local opt-out on Hits Radio UK. It went national (Fig. 4) on DAB and Freeview in June, appearing on over 20 extra local multiplexes.
Instead of moving to one of the national multiplex operators, however, the station is changing the branding of its existing The Hits channels, which are already on local multiplexes in the north of England and Scotland.
I would not be surprised if a similar rebranding were to happen to more of the founding stations of commercial radio such as Metro Radio in Newcastle, Radio City in Liverpool, Clyde 1 in Glasgow and Forth Radio in Edinburgh. These stations are part of Bauer’s Planet Radio group.
The group’s two stations, Key 103 and Key 2, are on the Manchester multiplex on block 12C, alongside The Hits. The station has been known as Key 103 for the past 30 years in Manchester. Before that, it was Piccadilly Radio, the city’s first ever commercial radio station; it launched in 1974.
In Manchester, the station will be called Hits Radio Manchester.
DAB News from Europe
The VRT First Layer multiplex operator in Belgium has allocated 96 kb/s capacity to all the music stations and 64kb/s capacity to the only speech station. Ten music stations are transmitting audio at 80 kb/s for audio and associated slideshows at 16kb/s. Just one station is using the full allocation for audio.
All the music stations are in DAB+ and are using the HE-AAC-V1 encoder. Had they opted for the V2 encoder, the audio would be about FM quality and that would be close to what DAB was supposed to achieve.
In northwest France, five of the seven new multiplexes should now be in operation in Dunkirk (8D), Calais (5B) and Lille (7C and 7D). There is a regional network on 8A. All networks use DAB+.
On the regional DAB + network, 13 stations will broadcast, including Skyrock and Sud Radio. The overall plan is that all towns and cities with more than 175,000 inhabitants will have DAB+ coverage by 2020.
Meanwhile, Project Orpheus, funded by the European Broadcasting Union (EBU, Fig. 5) is imagining a new approach to radio broadcasting. It is called object-based audio.
This is a software coding term and, in itself, does not tell us very much. In traditional broadcasting. a programme is made and transmitted in a linear way: This means that every listener on every device gets the same audio output.
In object-based broadcasting, the programme is turned into a collection of media files with metadata on how they should be reassembled in the receiver and sent to everyone.
In the receiver, objects can be re-assembled via metadata links or assembled differently by the user.
For example, take a football commentary: There will be the presenters, crowd noise near the pitch and at other points in the stadium, the referee’s audio and the players. These ‘audio objects’ can be adjusted in level, filtered or moved in the stereo image to create a personalized audio experience in your own home.
In this context, the project team is looking at advance audio protocols such as the 7.1+4H MPEG-H surround sound standard. This has seven speakers at ground level and four speakers overhead; The letter ‘H’ stands for ‘height’.
The DAB+ standard supports surround sound and upgrading to MPEG-H/AAC is straightforward.
Orpheus developed an app for iOS on the iPhone and, after three weeks, I received a link to redeem the app from Apple iTunes. The app provides a demo of the Orpheus idea and can be configured for various environments. It is still a little unstable.
Nevertheless, I enjoyed running the live feed, which provided a tutorial on the technology (Fig. 6)
RRI and DRM
Radio Romania International has used DRM for its broadcasts for many years but I find it very difficult to decode. The English service at 0530 UTC on 7330kHz is adjacent to an analogue signal, Radio Marti, on 7335kHz. There are plenty of free channels that time of the morning. However, those responsible for the transmissions, sometimes seem to plan their broadcasts without any understanding of DRM and interference.
It has to be said, however, that they are not the only organization doing this.
WINB on DRM
WINB, the longest-running shortwave broadcaster in the USA, is testing its new 15kW transmitter in the DRM mode, employing a hybrid mode that has not been used before. Initially, it looked like a simulcast mode meant for medium wave, where the analogue signal is in one 5kHz sideband and DRM in the other one.
I read the FCC Construction Permit and it looks like WINB may be testing some form of datacasting alongside DRM. Now that would be a very interesting digital radio development. The FCC documents refer to an emission type (G9W) which stands for a combination of analogue and digital channels.
Radio Kuwait DRM
If you want to try decoding DRM, the best option for beginners at the moment are the transmissions from Radio Kuwait to Europe (Fig. 7). I have mentioned the Arabic service before. It is broadcast from 0900 to 1325 UTC on 15110kHz. The English section of its multilingual programme is now strong from the start at 18.00 UTC on 15540kHz. Try to catch the service before 19.00 UTC; the signal starts fading after that.
I can’t recommend any DRM receivers at the moment as the only one in production is the Gospell GR-216 that will cost about £225 to buy directly from the manufacturer in China. The receiver costs £150 and shipping by air adds £75 to this.
Avion Electronics in India is working on a new version of its AV-DR-1401 model. I enquired if the firm would make this model available outside India and I have had no reply so far.
The latest news on the Titus II tablet based SDR (now a multi-standard digital receiver) is that it will be ready by the end of the year, after the firm has resolved issues with the HDRadio app for use in North America. The company seems to be planning to undertake a production run of 10,000 units.
SDR and Dream
There are two cheaper ways to receive DRM. The (not-so-popular) first option is to modify an analogue receiver to produce an IF of 12KHz that connects to a PC’s soundcard where software like Dream decodes the signal.
I did this years ago with a Sangean ATS803A and with the aid of a converter board from Sat-Schneider. The pdf sheets at these URLs will give you a good idea of what exactly is involved:
The second option is to use a cheap Software Defined Radio (SDR) using the RTL2832 chipset on your PC and to then connect the audio from that to Dream, by using a virtual audio cable. There is a good tutorial on this at the top of the RTL-SDR website. Remember, the device needs the R820T/T2 chip; you will also have to extend the frequency range down to about 100kHz.
Digital listening has exceeded 50% in the first quarter of this year, which is the last trigger for the Government to consider what to do about setting a digital switchover date.
My expectation is that the decision will be postponed.
Until next month, enjoy listening to digital radio in all its modes.
This article was featured in the August 2018 issue of Radio User